Fast and accurate phonetic spoken term detection

For the first time in human history, large volumes of spoken audio are being broadcast, made available on the internet, archived, and monitored for surveillance every day. New technologies are urgently required to unlock these vast and powerful stores of information. Spoken Term Detection (STD) systems provide access to speech collections by detecting individual occurrences of specified search terms. The aim of this work is to develop improved STD solutions based on phonetic indexing. In particular, this work aims to develop phonetic STD systems for applications that require open-vocabulary search, fast indexing and search speeds, and accurate term detection. Within this scope, novel contributions are made within two research themes, that is, accommodating phone recognition errors and, secondly, modelling uncertainty with probabilistic scores. A state-of-the-art Dynamic Match Lattice Spotting (DMLS) system is used to address the problem of accommodating phone recognition errors with approximate phone sequence matching. Extensive experimentation on the use of DMLS is carried out and a number of novel enhancements are developed that provide for faster indexing, faster search, and improved accuracy. Firstly, a novel comparison of methods for deriving a phone error cost model is presented to improve STD accuracy, resulting in up to a 33% improvement in the Figure of Merit. A method is also presented for drastically increasing the speed of DMLS search by at least an order of magnitude with no loss in search accuracy. An investigation is then presented of the effects of increasing indexing speed for DMLS, by using simpler modelling during phone decoding, with results highlighting the trade-off between indexing speed, search speed and search accuracy. The Figure of Merit is further improved by up to 25% using a novel proposal to utilise word-level language modelling during DMLS indexing. Analysis shows that this use of language modelling can, however, be unhelpful or even disadvantageous for terms with a very low language model probability. The DMLS approach to STD involves generating an index of phone sequences using phone recognition. An alternative approach to phonetic STD is also investigated that instead indexes probabilistic acoustic scores in the form of a posterior-feature matrix. A state-of-the-art system is described and its use for STD is explored through several experiments on spontaneous conversational telephone speech. A novel technique and framework is proposed for discriminatively training such a system to directly maximise the Figure of Merit. This results in a 13% improvement in the Figure of Merit on held-out data. The framework is also found to be particularly useful for index compression in conjunction with the proposed optimisation technique, providing for a substantial index compression factor in addition to an overall gain in the Figure of Merit. These contributions significantly advance the state-of-the-art in phonetic STD, by improving the utility of such systems in a wide range of applications.

Development of Contact Lenses from a Biomaterial Point of View – Materials, Manufacture, and Clinical Application

Rigid lenses, which were originally made from glass (between 1888 and 1940) and later from polymethyl methacrylate or silicone acrylate materials, are uncomfortable to wear and are now seldom fitted to new patients. Contact lenses became a popular mode of ophthalmic refractive error correction following the discovery of the first hydrogel material – hydroxyethyl methacrylate – by Czech chemist Otto Wichterle in 1960. To satisfy the requirements for ocular biocompatibility, contact lenses must be transparent and optically stable (for clear vision), have a low elastic modulus (for good comfort), have a hydrophilic surface (for good wettability), and be permeable to certain metabolites, especially oxygen, to allow for normal corneal metabolism and respiration during lens wear. A major breakthrough in respect of the last of these requirements was the development of silicone hydrogel soft lenses in 1999 and techniques for making the surface hydrophilic. The vast majority of contact lenses distributed worldwide are mass-produced using cast molding, although spin casting is also used. These advanced mass-production techniques have facilitated the frequent disposal of contact lenses, leading to improvements in ocular health and fewer complications. More than one-third of all soft contact lenses sold today are designed to be discarded daily (i.e., ‘daily disposable’ lenses).

Prediction of CO concentrations based on a hybrid Partial Least Square and Support Vector Machine model

Due to the health impacts caused by exposures to air pollutants in urban areas, monitoring and forecasting of air quality parameters have become popular as an important topic in atmospheric and environmental research today. The knowledge on the dynamics and complexity of air pollutants behavior has made artificial intelligence models as a useful tool for a more accurate pollutant concentration prediction. This paper focuses on an innovative method of daily air pollution prediction using combination of Support Vector Machine (SVM) as predictor and Partial Least Square (PLS) as a data selection tool based on the measured values of CO concentrations. The CO concentrations of Rey monitoring station in the south of Tehran, from Jan. 2007 to Feb. 2011, have been used to test the effectiveness of this method. The hourly CO concentrations have been predicted using the SVM and the hybrid PLS–SVM models. Similarly, daily CO concentrations have been predicted based on the aforementioned four years measured data. Results demonstrated that both models have good prediction ability; however the hybrid PLS–SVM has better accuracy. In the analysis presented in this paper, statistic estimators including relative mean errors, root mean squared errors and the mean absolute relative error have been employed to compare performances of the models. It has been concluded that the errors decrease after size reduction and coefficients of determination increase from 56 to 81% for SVM model to 65–85% for hybrid PLS–SVM model respectively. Also it was found that the hybrid PLS–SVM model required lower computational time than SVM model as expected, hence supporting the more accurate and faster prediction ability of hybrid PLS–SVM model.

Accurate urban road model reconstruction from high resolution remotely sensed imagery based on Support Vector Machine and Gabor filters

In this paper, we presented an automatic system for precise urban road model reconstruction based on aerial images with high spatial resolution. The proposed approach consists of two steps: i) road surface detection and ii) road pavement marking extraction. In the first step, support vector machine (SVM) was utilized to classify the images into two categories: road and non-road. In the second step, road lane markings are further extracted on the generated road surface based on 2D Gabor filters. The experiments using several pan-sharpened aerial images of Brisbane, Queensland have validated the proposed method.

Patient-specific computational biomechanics for simulating adolescent scoliosis surgery : predicted vs clinical correction for a preliminary series of six patients

Scoliosis is a spinal deformity that requires surgical correction in progressive cases. In order to optimize surgical outcomes, patient-specific finite element models are being developed by our group. In this paper, a single rod anterior correction procedure is simulated for a group of six scoliosis patients. For each patient, personalised model geometry was derived from low-dose CT scans, and clinically measured intra-operative corrective forces were applied. However, tissue material properties were not patient-specific, being derived from existing literature. Clinically, the patient group had a mean initial Cobb angle of 47.3 degrees, which was corrected to 17.5 degrees after surgery. The mean simulated post-operative Cobb angle for the group was 18.1 degrees. Although this represents good agreement between clinical and simulated corrections, the discrepancy between clinical and simulated Cobb angle for individual patients varied between -10.3 and +8.6 degrees, with only three of the six patients matching the clinical result to within accepted Cobb measurement error of +-5 degrees. The results of this study suggest that spinal tissue material properties play an important role in governing the correction obtained during surgery, and that patient-specific modelling approaches must address the question of how to prescribe patient-specific soft tissue properties for spine surgery simulation.

Design of a preliminary error impact analysis model for spatial safety assessment of earthmoving operations

Regardless of technology benefits, safety planners still face difficulties explaining errors related to the use of different technologies and evaluating how the errors impact the performance of safety decision making. This paper presents a preliminary error impact analysis testbed to model object identification and tracking errors caused by image-based devices and algorithms and to analyze the impact of the errors for spatial safety assessment of earthmoving and surface mining activities. More specifically, this research designed a testbed to model workspaces for earthmoving operations, to simulate safety-related violations, and to apply different object identification and tracking errors on the data collected and processed for spatial safety assessment. Three different cases were analyzed based on actual earthmoving operations conducted at a limestone quarry. Using the testbed, the impacts of the errors were investigated for the safety planning purpose.

Whole-proteome phylogeny of large dsDNA viruses and parvoviruses through a composition vector method related to dynamical language model

Background The vast sequence divergence among different virus groups has presented a great challenge to alignment-based analysis of virus phylogeny. Due to the problems caused by the uncertainty in alignment, existing tools for phylogenetic analysis based on multiple alignment could not be directly applied to the whole-genome comparison and phylogenomic studies of viruses. There has been a growing interest in alignment-free methods for phylogenetic analysis using complete genome data. Among the alignment-free methods, a dynamical language (DL) method proposed by our group has successfully been applied to the phylogenetic analysis of bacteria and chloroplast genomes. Results In this paper, the DL method is used to analyze the whole-proteome phylogeny of 124 large dsDNA viruses and 30 parvoviruses, two data sets with large difference in genome size. The trees from our analyses are in good agreement to the latest classification of large dsDNA viruses and parvoviruses by the International Committee on Taxonomy of Viruses (ICTV). Conclusions The present method provides a new way for recovering the phylogeny of large dsDNA viruses and parvoviruses, and also some insights on the affiliation of a number of unclassified viruses. In comparison, some alignment-free methods such as the CV Tree method can be used for recovering the phylogeny of large dsDNA viruses, but they are not suitable for resolving the phylogeny of parvoviruses with a much smaller genome size.

Model selection and error estimation

We study model selection strategies based on penalized empirical loss minimization. We point out a tight relationship between error estimation and data-based complexity penalization: any good error estimate may be converted into a data-based penalty function and the performance of the estimate is governed by the quality of the error estimate. We consider several penalty functions, involving error estimates on independent test data, empirical VC dimension, empirical VC entropy, and margin-based quantities. We also consider the maximal difference between the error on the first half of the training data and the second half, and the expected maximal discrepancy, a closely related capacity estimate that can be calculated by Monte Carlo integration. Maximal discrepancy penalty functions are appealing for pattern classification problems, since their computation is equivalent to empirical risk minimization over the training data with some labels flipped.

Fast rates for estimation error and oracle inequalities for model section

We consider complexity penalization methods for model selection. These methods aim to choose a model to optimally trade off estimation and approximation errors by minimizing the sum of an empirical risk term and a complexity penalty. It is well known that if we use a bound on the maximal deviation between empirical and true risks as a complexity penalty, then the risk of our choice is no more than the approximation error plus twice the complexity penalty. There are many cases, however, where complexity penalties like this give loose upper bounds on the estimation error. In particular, if we choose a function from a suitably simple convex function class with a strictly convex loss function, then the estimation error (the difference between the risk of the empirical risk minimizer and the minimal risk in the class) approaches zero at a faster rate than the maximal deviation between empirical and true risks. In this paper, we address the question of whether it is possible to design a complexity penalized model selection method for these situations. We show that, provided the sequence of models is ordered by inclusion, in these cases we can use tight upper bounds on estimation error as a complexity penalty. Surprisingly, this is the case even in situations when the difference between the empirical risk and true risk (and indeed the error of any estimate of the approximation error) decreases much more slowly than the complexity penalty. We give an oracle inequality showing that the resulting model selection method chooses a function with risk no more than the approximation error plus a constant times the complexity penalty.

The Bayesian conditional independence model for measurement error: applications in ecology

The measurement error model is a well established statistical method for regression problems in medical sciences, although rarely used in ecological studies. While the situations in which it is appropriate may be less common in ecology, there are instances in which there may be benefits in its use for prediction and estimation of parameters of interest. We have chosen to explore this topic using a conditional independence model in a Bayesian framework using a Gibbs sampler, as this gives a great deal of flexibility, allowing us to analyse a number of different models without losing generality. Using simulations and two examples, we show how the conditional independence model can be used in ecology, and when it is appropriate.

Correction of step artefact associated with MRI scanning of long bones

3D models of long bones are being utilised for a number of fields including orthopaedic implant design. Accurate reconstruction of 3D models is of utmost importance to design accurate implants to allow achieving a good alignment between two bone fragments. Thus for this purpose, CT scanners are employed to acquire accurate bone data exposing an individual to a high amount of ionising radiation. Magnetic resonance imaging (MRI) has been shown to be a potential alternative to computed tomography (CT) for scanning of volunteers for 3D reconstruction of long bones, essentially avoiding the high radiation dose from CT. In MRI imaging of long bones, the artefacts due to random movements of the skeletal system create challenges for researchers as they generate inaccuracies in the 3D models generated by using data sets containing such artefacts. One of the defects that have been observed during an initial study is the lateral shift artefact occurring in the reconstructed 3D models. This artefact is believed to result from volunteers moving the leg during two successive scanning stages (the lower limb has to be scanned in at least five stages due to the limited scanning length of the scanner). As this artefact creates inaccuracies in the implants designed using these models, it needs to be corrected before the application of 3D models to implant design. Therefore, this study aimed to correct the lateral shift artefact using 3D modelling techniques. The femora of five ovine hind limbs were scanned with a 3T MRI scanner using a 3D vibe based protocol. The scanning was conducted in two halves, while maintaining a good overlap between them. A lateral shift was generated by moving the limb several millimetres between two scanning stages. The 3D models were reconstructed using a multi threshold segmentation method. The correction of the artefact was achieved by aligning the two halves using the robust iterative closest point (ICP) algorithm, with the help of the overlapping region between the two. The models with the corrected artefact were compared with the reference model generated by CT scanning of the same sample. The results indicate that the correction of the artefact was achieved with an average deviation of 0.32 ± 0.02 mm between the corrected model and the reference model. In comparison, the model obtained from a single MRI scan generated an average error of 0.25 ± 0.02 mm when compared with the reference model. An average deviation of 0.34 ± 0.04 mm was seen when the models generated after the table was moved were compared to the reference models; thus, the movement of the table is also a contributing factor to the motion artefacts.

Exponential integrators and a dual-scale model for wood drying

For the timber industry, the ability to simulate the drying of wood is invaluable for manufacturing high quality wood products. Mathematically, however, modelling the drying of a wet porous material, such as wood, is a diffcult task due to its heterogeneous and anisotropic nature, and the complex geometry of the underlying pore structure. The well{ developed macroscopic modelling approach involves writing down classical conservation equations at a length scale where physical quantities (e.g., porosity) can be interpreted as averaged values over a small volume (typically containing hundreds or thousands of pores). This averaging procedure produces balance equations that resemble those of a continuum with the exception that effective coeffcients appear in their deffnitions. Exponential integrators are numerical schemes for initial value problems involving a system of ordinary differential equations. These methods differ from popular Newton{Krylov implicit methods (i.e., those based on the backward differentiation formulae (BDF)) in that they do not require the solution of a system of nonlinear equations at each time step but rather they require computation of matrix{vector products involving the exponential of the Jacobian matrix. Although originally appearing in the 1960s, exponential integrators have recently experienced a resurgence in interest due to a greater undertaking of research in Krylov subspace methods for matrix function approximation. One of the simplest examples of an exponential integrator is the exponential Euler method (EEM), which requires, at each time step, approximation of φ(A)b, where φ(z) = (ez - 1)/z, A E Rnxn and b E Rn. For drying in porous media, the most comprehensive macroscopic formulation is TransPore [Perre and Turner, Chem. Eng. J., 86: 117-131, 2002], which features three coupled, nonlinear partial differential equations. The focus of the first part of this thesis is the use of the exponential Euler method (EEM) for performing the time integration of the macroscopic set of equations featured in TransPore. In particular, a new variable{ stepsize algorithm for EEM is presented within a Krylov subspace framework, which allows control of the error during the integration process. The performance of the new algorithm highlights the great potential of exponential integrators not only for drying applications but across all disciplines of transport phenomena. For example, when applied to well{ known benchmark problems involving single{phase liquid ow in heterogeneous soils, the proposed algorithm requires half the number of function evaluations than that required for an equivalent (sophisticated) Newton{Krylov BDF implementation. Furthermore for all drying configurations tested, the new algorithm always produces, in less computational time, a solution of higher accuracy than the existing backward Euler module featured in TransPore. Some new results relating to Krylov subspace approximation of '(A)b are also developed in this thesis. Most notably, an alternative derivation of the approximation error estimate of Hochbruck, Lubich and Selhofer [SIAM J. Sci. Comput., 19(5): 1552{1574, 1998] is provided, which reveals why it performs well in the error control procedure. Two of the main drawbacks of the macroscopic approach outlined above include the effective coefficients must be supplied to the model, and it fails for some drying configurations, where typical dual{scale mechanisms occur. In the second part of this thesis, a new dual{scale approach for simulating wood drying is proposed that couples the porous medium (macroscale) with the underlying pore structure (microscale). The proposed model is applied to the convective drying of softwood at low temperatures and is valid in the so{called hygroscopic range, where hygroscopically held liquid water is present in the solid phase and water exits only as vapour in the pores. Coupling between scales is achieved by imposing the macroscopic gradient on the microscopic field using suitably defined periodic boundary conditions, which allows the macroscopic ux to be defined as an average of the microscopic ux over the unit cell. This formulation provides a first step for moving from the macroscopic formulation featured in TransPore to a comprehensive dual{scale formulation capable of addressing any drying configuration. Simulation results reported for a sample of spruce highlight the potential and flexibility of the new dual{scale approach. In particular, for a given unit cell configuration it is not necessary to supply the effective coefficients prior to each simulation.

Evaluation of a patient-specific finite-element model to simulate conservative treatment in adolescent idiopathic scoliosis

Study design Retrospective validation study. Objectives To propose a method to evaluate, from a clinical standpoint, the ability of a finite-element model (FEM) of the trunk to simulate orthotic correction of spinal deformity and to apply it to validate a previously described FEM. Summary of background data Several FEMs of the scoliotic spine have been described in the literature. These models can prove useful in understanding the mechanisms of scoliosis progression and in optimizing its treatment, but their validation has often been lacking or incomplete. Methods Three-dimensional (3D) geometries of 10 patients before and during conservative treatment were reconstructed from biplanar radiographs. The effect of bracing was simulated by modeling displacements induced by the brace pads. Simulated clinical indices (Cobb angle, T1–T12 and T4–T12 kyphosis, L1–L5 lordosis, apical vertebral rotation, torsion, rib hump) and vertebral orientations and positions were compared to those measured in the patients' 3D geometries. Results Errors in clinical indices were of the same order of magnitude as the uncertainties due to 3D reconstruction; for instance, Cobb angle was simulated with a root mean square error of 5.7°, and rib hump error was 5.6°. Vertebral orientation was simulated with a root mean square error of 4.8° and vertebral position with an error of 2.5 mm. Conclusions The methodology proposed here allowed in-depth evaluation of subject-specific simulations, confirming that FEMs of the trunk have the potential to accurately simulate brace action. These promising results provide a basis for ongoing 3D model development, toward the design of more efficient orthoses.

Robust estimating functions and bias correction for longitudinal data analysis

Robust methods are useful in making reliable statistical inferences when there are small deviations from the model assumptions. The widely used method of the generalized estimating equations can be "robustified" by replacing the standardized residuals with the M-residuals. If the Pearson residuals are assumed to be unbiased from zero, parameter estimators from the robust approach are asymptotically biased when error distributions are not symmetric. We propose a distribution-free method for correcting this bias. Our extensive numerical studies show that the proposed method can reduce the bias substantially. Examples are given for illustration.

Spectral unmixing with estimated adaptive endmember index using Extended Support Vector Machine

The most difficult operation in the flood inundation mapping using optical flood images is to separate fully inundated areas from the ‘wet’ areas where trees and houses are partly covered by water. This can be referred as a typical problem the presence of mixed pixels in the images. A number of automatic information extraction image classification algorithms have been developed over the years for flood mapping using optical remote sensing images. Most classification algorithms generally, help in selecting a pixel in a particular class label with the greatest likelihood. However, these hard classification methods often fail to generate a reliable flood inundation mapping because the presence of mixed pixels in the images. To solve the mixed pixel problem advanced image processing techniques are adopted and Linear Spectral unmixing method is one of the most popular soft classification technique used for mixed pixel analysis. The good performance of linear spectral unmixing depends on two important issues, those are, the method of selecting endmembers and the method to model the endmembers for unmixing. This paper presents an improvement in the adaptive selection of endmember subset for each pixel in spectral unmixing method for reliable flood mapping. Using a fixed set of endmembers for spectral unmixing all pixels in an entire image might cause over estimation of the endmember spectra residing in a mixed pixel and hence cause reducing the performance level of spectral unmixing. Compared to this, application of estimated adaptive subset of endmembers for each pixel can decrease the residual error in unmixing results and provide a reliable output. In this current paper, it has also been proved that this proposed method can improve the accuracy of conventional linear unmixing methods and also easy to apply. Three different linear spectral unmixing methods were applied to test the improvement in unmixing results. Experiments were conducted in three different sets of Landsat-5 TM images of three different flood events in Australia to examine the method on different flooding conditions and achieved satisfactory outcomes in flood mapping.